ComfyUI Node
Nested Detailer
Detail faces and then eyes within the face. FastDetailer nodes always scale regions to 1MP so this does the nested detailing on the scaled image instead of doing it on the final composited image.
Nested Detailer
- image
- model
- vae
- positive
- negative
- image
- detailed_faces
◄face_model▾►
◄eyes_pair_model▾►
◄eye_single_model▾►
◄threshold0.50►
◄seed0►
◄cfg1.5►
◄samplereuler_ancestral_cfg_pp►
◄scheduleralign_your_steps►
◄face_steps20►
◄face_denoise0.40►
◄face_scale1.5►
◄eye_steps20►
◄eye_denoise0.40►
◄eye_scale1.5►
◄upscale_method▾►
◄max_megapixels1.5►
◄feather0.20►
◄context_padding0.10►
Categorymudknight/detailer
Inputs (23)
| Name | Type | Default | Description |
|---|---|---|---|
| face_model | COMBO | 1 options: bbox/face_yolov8m.pt | |
| eyes_pair_model | COMBO | 1 options: bbox/face_yolov8m.pt | |
| eye_single_model | COMBO | 1 options: bbox/face_yolov8m.pt | |
| threshold | FLOAT | 0.500–1 | Detection threshold for all models |
| image | IMAGE | — | |
| model | MODEL | — | |
| vae | VAE | — | |
| positive | CONDITIONING | — | |
| negative | CONDITIONING | — | |
| seed | INT | 00–18446744073709550000 | — |
| cfg | FLOAT | 1.50–100 | — |
| sampler | COMBO | euler_ancestral_cfg_pp | 44 options: euler, euler_cfg_pp, euler_ancestral, euler_ancestral_cfg_pp, heun, heunpp2, +38 |
| scheduler | COMBO | align_your_steps | 10 options: simple, sgm_uniform, karras, exponential, ddim_uniform, beta, +4 |
| face_steps | INT | 201–10000 | — |
| face_denoise | FLOAT | 0.400–1 | — |
| face_scale | FLOAT | 1.50.1–2 | Amount to upscale cropped region before sampling. Higher values create more detailed images. |
| eye_steps | INT | 201–10000 | — |
| eye_denoise | FLOAT | 0.400–1 | — |
| eye_scale | FLOAT | 1.50.1–2 | Amount to upscale cropped region before sampling. Higher values create more detailed images. |
| upscale_method | COMBO | 5 options: lanczos, bilinear, bicubic, area, nearest-exact | |
| max_megapixels | FLOAT | 1.50.1–10 | Maximum size in megapixels for upscaled images |
| feather | FLOAT | 0.200–1 | Percentage of image to feather when uncropping |
| context_padding | FLOAT | 0.100–1 | Percentage of image to use for context from edge |
Outputs (2)
| Name | Type | Description |
|---|---|---|
| image | IMAGE | — |
| detailed_faces | IMAGE | — |